The confident call that the basket disproved
An experienced operator at an Indian packaged-food business recently concluded that a new snack line should be priced and flavoured for metro households because that was where premium consumption had grown fastest in his previous role. He expected a familiar national flavour to lead, with a higher price justified by better ingredients. The first purchase data told a different story. Orders from smaller cities produced a larger share of repeat purchases, and the leading flavour was a regional variant that the team had treated as a secondary option. The lower price tier also converted more consistently than the premium pack.
This is a plausible pattern, not an unusual failure of judgement. The operator knew the category, understood distribution, and had seen several successful launches. His mistake was not a lack of experience. It was applying experience formed in an earlier market to a current market with different channels, price pressures, and shopping habits.
Category intuition remains useful. It helps a team form an initial hypothesis, spot an implausible result, and recognise consumer language that a spreadsheet may miss. It becomes dangerous when a strong prior is treated as evidence, especially before a brand has observed who buys, what they choose, and whether they return.
Old category wisdom meets a more uneven India
Indian consumer categories are difficult to read from averages. A national number can conceal sharp differences between metros and smaller cities, modern trade and marketplaces, first-time trial and repeat buying. A product that appears premium in a large city may be considered an accessible upgrade elsewhere. A flavour that feels broadly familiar to a team in Mumbai or Bengaluru may have limited relevance in another region.
E-commerce makes these differences easier to see, but not automatically easier to interpret. A digital storefront can reach consumers outside the cities that shaped a team's assumptions. Promotions can bring in buyers who would not have entered the category at full price. Delivery coverage and platform placement can change the mix of customers from one month to the next.
That means accepted category wisdom needs a date and a context. "Consumers here will not pay for that," "this category is driven by gifting," or "the national flavour is the safe launch choice" may once have been reasonable observations. They are not permanent laws. Purchase behaviour can shift before industry commentary catches up.
Where purchase data contradicts the story
Consider a synthetic example from personal care. A team with years of experience in urban retail may believe that a smaller, premium pack will win online because shoppers want to trial a new product without committing to a larger format. They launch with refined packaging, a narrow set of variants, and a price positioned above mass alternatives.
After several weeks, the purchase data shows that the larger value pack is selling more often among households outside the largest cities. The buyers are not merely responding to a discount. They are purchasing for more than one person, comparing the effective price per use, and reordering when the product fits an established routine. A regional fragrance may also outperform the neutral option that the team considered universally acceptable.
The right response is not to declare intuition useless or to follow every data point blindly. The team should check whether the sample is being shaped by promotions, stock availability, delivery reach, or a particular platform. It should then compare conversion, repeat rate, pack preference, geography, and contribution by cohort. The point is to replace a confident generalisation with a testable explanation.
In another category, a founder might assume that a higher-priced health product will attract the most committed users because that is how the category behaved in specialist stores. Online purchase data may show a different path: first orders arrive through entry packs, while repeat buyers gradually move to larger formats. The initial price is influencing trial, but the repeat purchase is revealing willingness to pay. Without separating those stages, a team can mistake a trial barrier for a demand ceiling.
Building institutional data discipline
Data discipline does not require a large analytics department. It starts with recording the decisions that matter before the result arrives. For each product or test, write down the expected buyer, price, pack, channel, region, acquisition source, and likely reason for repeat. Then define what would count as evidence against the hypothesis.
A small team can review a short operating set each week:
- Orders and conversion by channel, region, pack, and price tier
- First purchase compared with second and third purchase behaviour
- Repeat timing, refund or complaint signals, and stock-out effects
- Variant-level performance, including regional preferences
- Customer acquisition cost and contribution after discounts and fulfilment
The discipline lies in keeping the definitions stable. If "repeat" means a second order within one period this week and any later order next week, the trend is not comparable. If a promotion is included in one cohort but ignored in another, price sensitivity will be misread. Simple, consistent records are more valuable than a complicated dashboard that nobody trusts.
It also helps to assign a decision owner. Someone should be responsible for stating what the data says, someone should challenge the interpretation, and the team should record what it will change. For a three-person company, these can be rotating responsibilities rather than separate roles. The goal is to make learning survive beyond the memory of the person who ran the test.
When intuition still earns its place
Category intuition is more reliable when the conditions are stable and the operator has direct, recent exposure to the same customer and channel. A well-understood regional preference, a regulated usage pattern, or a mature category with predictable replenishment can provide a strong starting point. In those cases, experience may help a team avoid unnecessary tests and recognise a meaningful deviation quickly.
Even then, intuition should set the question rather than close the discussion. An operator who knows that a region prefers a particular taste can make that the lead hypothesis. Purchase data can show whether the preference survives the actual price, pack, channel, and product experience. The combination is stronger than either source alone.
At Palette Brands, our three-person team uses this standard while evaluating Indian consumer opportunities. In our early incubation program, internal pilot tracking across three Indian consumer categories showed a 58% average reorder rate across products, compared with a 31% category baseline. This is internal pilot framing, not a claim about the wider market. Its value is that it keeps the conversation anchored in observed purchasing and reordering rather than category reputation.
Institutional data discipline for a small team means writing down the belief, measuring the purchase, checking the customer context, and making the next decision traceable. Founders and operators can run their category numbers with Palette to check their intuition against the data.
